Prostate cancer is an extremely prevalent disease, resulting in the deaths of over 41,800 men annually in the United States. When prostate cancer is confined to the gland, or is minimally invasive into the seminal vesicles and capsule, it has been conventionally treated by surgical removal. An increasing number of patients are now selecting focal radiation therapies (conformal radiation therapy and Brachytherapy) which have demonstrated lower complication rates without survival disadvantage, and are much less intrusive. Selection of appropriate patients, therapeutic planning, and assessment of therapeutic efficacy of focal radiation therapies requires an accurate knowledge of the location, spatial extent and aggressiveness of the cancer prior to and after therapy. Currently available clinical measures (PSA, histology of biopsies) and non-invasive radiologic techniques (transrectal ultrasound, CT, MRI) often cannot reliably provide this information. The goal of the current proposal is to determine whether the metabolic information provided by 3-dimensional 1H spectroscopic imaging (MRSI) can accurately assess the spatial extent and grade of prostate cancer prior to and after conformal radiation therapy and radioactive seed implantation (Brachytherapy). The feasibility of accomplishing this goal is supported by our experience in performing over 800 MRI/MRSI examinations of prostate cancer patients, the large patient population at UCSF, the direct involvement of radiation oncologists and urologists in the project, and the large amount of preliminary MRSI/MRI data which support the biochemical hypotheses which form the basis of this proposal. Specifically, in a 3-D MRSI study of 85 prostate cancer patients prior to radical resection, we demonstrated that choline and citrate levels could assess the presence and spatial extent of prostate cancer prior to therapy with high specificity. In a recently completed study of 62 pre-radical prostatectomy patients with step section histopathology, we demonstrated for the first time that the addition of MRSI to MRI can significantly improve both cancer staging and localization of cancer within the prostate prior to therapy. There is also preliminary evidence that the magnitude of the changes in choline and citrate are correlated with cancer grade, and that MRSI can discriminate cancer from necrosis and other viable tissues after therapy. However, there has not been to date, a serial prospective MRSI study on a clinically significant number of prostate cancer patients receiving radiotherapy. The data acquired in this proposal should allow us to determine if combined MRI/MRSI can: 1) improve our understanding of the metabolic effects of radiation therapy on normal and malignant prostatic tissues, 2) improve the selection of patients for radiation therapy, 3) aid in therapeutic planning, and 4) allow a early, more quantitative assessment of therapeutic efficacy.